Agent overview (udiagent)

LLM-powered data visualization orchestration library for the Universal Discovery Interface (UDI).

UDIAgent orchestrates LLM calls to generate data visualization specs from natural language queries. It can be used as a standalone Python library or deployed as a FastAPI microservice.

Installation

# Core library only
pip install udiagent

# With the reference FastAPI server
pip install udiagent[server]

# With LangFuse observability
pip install udiagent[langfuse]

# With Amazon Bedrock IAM-role (SigV4) authentication
pip install udiagent[bedrock]

# With benchmarking tools
pip install udiagent[benchmark]

# Everything
pip install udiagent[all]

For local development with uv:

uv sync --extra server --extra langfuse --extra test --extra bedrock   # server + dev

Quick start

As a library:

from udiagent import UDIAgent, Orchestrator

agent = UDIAgent(gpt_model_name="gpt-5.4", openai_api_key="sk-...")
result = Orchestrator(agent).run(
    messages=[{"role": "user", "content": "Show me a bar chart of donors by sex"}],
    data_schema='{"resources": [...]}',
    data_domains='[{"entity": "donors", "field": "sex", ...}]',
)
# result.tool_calls: the tool calls to apply (RenderVisualization, FilterData, ...)

As a server, for local development (from packages/agent):

INSECURE_DEV_MODE=1 uv run --extra server fastapi dev src/udiagent/server/app.py --port 8007

Documentation

Library usage: configuration, template sets, reasoning models, OpenAI-compatible backends, Amazon Bedrock, LangFuse.
Running the server: environment variables, identity providers, server-side query backends, Docker.
Deployment guide: standing the server up on a host.
How the agent works: the request flow, chart templates, filters, and regenerating the generated template files.
Benchmarking: measuring a server configuration.
Server-side data: compiling the grammar to SQL.
Reference: the Python API and the HTTP API (from openapi.json).

License

MIT